© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 24-TRANSCODER-HP
    7. 150629
    Prev
    Next
    INDEX
    Explanations

    parts of something

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_24.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.24.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ÅŁim
    -0.30
    oyal
    -0.27
    otty
    -0.26
    lates
    -0.24
    andes
    -0.24
    lasses
    -0.24
    å°ıäºĭ
    -0.24
    è¾¹å¢ĥ
    -0.24
    rightness
    -0.23
    缸è¿ij
    -0.23
    POSITIVE LOGITS
    Parts
    0.62
     Parts
    0.61
     parts
    0.60
    parts
    0.57
    ç»ĦæĪIJéĥ¨åĪĨ
    0.54
     components
    0.50
    ç»ĵæŀĦ
    0.50
     structure
    0.49
    _parts
    0.49
     partes
    0.46
    Activations Density 0.972%

    No Known Activations